Preliminary investigation of thermostable DNA polymerases to reduce PCR amplification artifacts

Forensic genotyping uses a multiplex short tandem repeat (STR) assay to amplify deoxyribonucleic acid (DNA) samples. One of the artifacts mostly commonly encountered in forensic DNA analysis is stutter, which are non-specific products from the polymerase chain reaction (PCR) that are typically one repeat unit shorter in length than the allelic amplicon. While stutter peaks are typically no taller than 10% of the parent peak on electropherograms, their peak heights can fall into similar ranges as minor contributor alleles in mixtures, creating a problem of how to distinguish artifacts from true allele peaks in these situations. One way to potentially address this issue is to find a PCR method that produces a much lower amount of stutter than the method currently used, which involves amplifying samples with commercial PCR kits designed for forensic applications. These kits all use some form of Taq DNA polymerase (derived from Thermus aquaticus). In an effort to examine whether the type of enzyme used in an assay affects the resulting stutter rates observed, the existing GlobalFiler™ PCR Amplification Kit (Applied Biosystems) protocol for forensic multiplex STR assays was modified to test different types of enzymes. This was done by amplifying the same DNA sample with GlobalFiler primers and different commercial proofreading enzymes and their accompanying reaction buffer using manufacturer-recommended PCR parameters. The DNA sample originated from a buccal swab that was extracted on the EZ1® Advanced (Qiagen). The DNA solution was quantified using the Quantifiler™ Duo DNA Quantification Kit (Applied Biosystems) on the 7500 Real-Time PCR System (Applied Biosystems). In order to first establish the validity of switching out enzymes in an established protocol, a DNA sample was amplified with the Type-it® Microsatellite Kit (Qiagen), another Taq-based kit that is also marketed for use in multiplex STR assays. After a complete profile was successfully generated, research proceeded with testing various high-fidelity DNA polymerases. Some of the enzymes tested were known to be Pyrococcus-like while others were fused to a DNA-binding domain to enhance processivity. Taq polymerases tend to produce products with 3’adenine-overhangs while proofreading enzymes produce blunt-ends. This change caused a one base pair difference in the resulting amplicon lengths, which was accommodated by manually assigning genotypes after results from fragment analysis by capillary electrophoresis using a 3130 Genetic Analyzer (Applied Biosystems) were interpreted by the GeneMapper™ software (Applied Biosystems). Additional amplification kits tested were: the UCP HiFidelity PCR Kit (Qiagen), Phusion™ Hot Start II High-Fidelity DNA Polymerase (Thermo Scientific), Platinum™ SuperFi™ II DNA Polymerase (Invitrogen), iProof™ High-Fidelity DNA Polymerase (Bio-Rad), Q5® High-Fidelity DNA Polymerase (New England Biolabs), and TruFi™ DNA Polymerase (Azura Genomics). Most of the kits produced profiles exhibiting a high degree of uneven amplification and varying levels of allelic dropout. In addition, all of the kits tested had much shorter peak heights compared to using GlobalFiler. Changing the type of enzyme used in an established protocol was found to be less straightforward than anticipated. Due to the poor quality results obtained in the first pass of trials, a few kits were selected to undergo optimization in the hopes of achieving higher quality results from which further analyses, such as comparing stutter rates, could be more reliably conducted. Both altered reagent amounts (higher enzyme concentration, higher DNA input mass) and different PCR parameters (decreased denaturation temperature, varying annealing temperature, decreased extension temperature, longer extension cycles, and longer final extension stage) were assessed. Only an increase in extension cycling time was found to produce better peak heights while maintaining balanced amplification of most of the targeted loci. Initial samples amplified with the Phusion enzyme exhibited multiple non-specific artifacts that were not stutter. Raising the annealing temperature for that enzyme’s protocol eliminated this issue. Therefore, higher annealing temperatures were pre-emptively used for several of the other enzymes tested. One of the explanations proposed for the uneven amplification observed is the presence of inhibitors in the commercial buffers used affecting downstream capillary electrophoresis. The Q5 High-Fidelity and TruFi DNA polymerases produced the best quality profiles; the UCP HiFidelity PCR Kit had the poorest results. Preliminary results indicated that none of the protocol alterations implemented significantly decreased stutter rates, nor was any one commercial enzyme found to have consistently lower stutter rates than the GlobalFiler kit. Due to the low number of trials carried out, the findings from this study require more replications with a wider variety of DNA polymerases to confirm that the type of enzyme used in an assay does not affect stutter rates